This is a longitudinal study of hormones and symptoms to characterize the fluctuations and rates of change of DHEAS, FSH, LH and estradiol and test hypotheses that 1) symptoms are associated with changes in these hormones; 2) DHEAS may have a direct effect on psychological symptoms; 3) psychological symptoms are not correlated with and occur prior to vasomotor symptoms; and 4) any differences in symptoms and hormones between African-American and Caucasian women is negligible after controlling for background variables known to differentiate these groups. The population-based sample has 400 women (200 African-American, 200 Caucasian), ages 35-47 at enrollment and is followed at 9-month intervals for 5 years. The study will provide important information that is now know about relationships between symptoms and hormones.